What Makes Transfer Learning Work For Medical Images: Feature Reuse & Other Factors.
Christos MatsoukasJohan Fredin HaslumMoein SorkheiMagnus SöderbergKevin SmithPublished in: CoRR (2022)
Keyphrases
- medical images
- transfer learning
- anatomical structures
- learning tasks
- knowledge transfer
- medical image analysis
- medical imaging
- labeled data
- deformable models
- cross domain
- medical image segmentation
- reinforcement learning
- active learning
- mr images
- collaborative filtering
- domain adaptation
- biomedical images
- text classification
- machine learning
- machine learning algorithms
- learning algorithm
- transfer knowledge
- segmentation of medical images
- medical image retrieval
- semi supervised learning
- target domain
- text categorization
- image features
- data mining
- data sets